Type III interferons, IL-28 and IL-29, are increased in chronic HCV infection and induce myeloid dendritic cell-mediated FoxP3+ regulatory T cells

Abstract

Background & aims: Hepatitis C virus (HCV) is difficult to eradicate and type III interferons (IFN-λ, composed of IL-28A, IL-28B and IL-29) are novel therapeutic candidates. We hypothesized that IFN-λ have immunomodulatory effects in HCV- infected individuals.

Materials and methods: We analyzed the expression of IFN-λ and its receptor (composed of IL-10R2 and IFN-λR subunits) in the blood and livers of patients with chronic (c)HCV infection compared to controls (those who cleared HCV by sustained virological response, SVR, and those with liver inflammation of non-viral origin, non-alcoholic steatohepatitis, NASH). We also compared the proliferative capacity of dendritic cells (DCs) obtained from healthy individuals and those with chronic HCV using a mixed leukocyte reaction combined with 3H-Td incorporation. In addition, the composition of the IFN-λ receptor (IFN-λR) on myeloid DCs, plasmacytoid DCs, PBMCs, and T cells was determined by FACS analysis.

Results: We report that the expression of IFN-λ protein in serum and mRNA in liver is increased in cHCV patients, but not in those with HCV SVR or NASH, compared to controls. Liver level of IFN-λR mirrored the expression of serum IFN-λ and was higher in cHCV, compared to controls and HCV-SVR patients, suggesting that elevation of IFN-λ and IFN-λR are HCV-dependent. We further identified that innate immune cell populations expressed complete IFN-λ receptor. In vitro, recombinant IFN-λ promoted differentiation of monocyte-derived dendritic cells (DCs) into a phenotype with low T cell stimulatory capacity and high PD-L1 expression, which further promoted expansion of existing regulatory T cells. IFN-λ-DCs failed to induce de novo generation of regulatory T cells. The inhibitory capacity of IFN-λ-DCs was counteracted by recombinant IL-12 and by neutralization of the PD-1/PD-L1 system.

Conclusions: Our novel findings of the immunomodulatory effect of IFN-λ contribute to the understanding of the anti-inflammatory and/or anti-viral potential of IFN-λ in cHCV.

Figure 1. HCV infection correlates with high levels of IFN-λ and IFN-λR.

Serum (A,B) and liver levels (C–E) of IL-28 (A-protein; C-mRNA), IL-29 (B-protein, D-mRNA), and IFN-λR (E, mRNA). There were 18 patients with chronic HCV infection, 12 controls, 16 HCV SVR and 12 NASH analyzed for serum cytokines (panels A,B). There were 18 patients with chronic HCV infection, 4 controls, and 16 SVR analyzed for RNA (panels C–E). The samples from blood (A,B) and liver RNA (C–E) were paired for HCV and SVR. (F) IFN-λR1 expression was analyzed in various cell types of PBMCs by flow cytometry using isotype or IFN-λR1 fluorescent antibody. Representative histogram of n = 4 is shown. * indicates p<0.05 compared to controls.

Figure 2. IFN-λ influences the DCs phenotype and functional capacity.

Monocyte-derived DCs from healthy individuals were generated without (control) or with a combination of IL-29 and IL-28A (control+IFN-λ), or from patients with chronic HCV (HCV+IFN-λ) or recovered HCV infection (SVR+IFN-λ) for 7 days, after which they were included in a mixed lymphocyte reaction (MLR) with allogeneic CD4⁺ T cells for 4 days. Incorporation of ³H-Td during the last 16 hours of MLR, read as counts per minute (cpm), was considered indicative of the stimulatory capacity of DCs (A, left). DCs of controls, HCV and SVR individuals were analyzed as above (A, right). Control DCs, DCs generated for 7 days with IFN-λ, and DCs exposed to IFN-λ only during the last 24 hrs on day 6, were included in MLR. Incorporation of ³H-Td during the last 16 hours of MLR, read as cpm, was considered indicative of stimulatory capacity of DCs (B). Normal monocytes were transfected with scrambled or IFN-λR-specific siRNA and analyzed for the content of IFN-λR-coding RNA 72 hours later by RT-qPCR, as described in Methods. Fold change compared to mock-transfected cells was calculated from n = 3 sets of experiments (C). Normal monocytes were transfected as above, exposed to IL-4⁺GM-CSF in the presence or absence of IFN-λ for 7 days and included in MLR with normal T cells. T cells alone and DCs alone were used as controls of non-specific proliferation. Incorporation of ³H-Td during the last 16 hours of MLR, read as cpm, was measured as indicative of stimulatory capacity of DCs from n = 5 (D). Total RNA was extracted from MLR co-cultures corresponding to those described in panel D and the content of FoxP3-coding mRNA was analyzed using RT-qPCR, as described in methods. Fold change compared to scrambled MLR which used siRNA-exposed DCs as control was calculated from n = 5 sets of experiments (E). T cells of control healthy individuals were treated with IFN-λ for 5 days, after which they were washed extensively and co-cultured in MLR with dendritic cells which were generated with IL-4⁺GM-CSF without IFN-λ (IFN-λ-naïve) for 4 days. There was no IFN-λ present during MLR. Incorporation of ³H-Td during the last 16 hours of MLR, read as cpm, was measured as indicative of stimulatory capacity of DCs from n = 5 sets of experiments (F). * indicates p<0.05 compared to controls.

Figure 3. IFN-λ treatment-induced inhibitory DCs phenotype is dependent on IL-10 and PD-1/PD-L1 ligand.

Monocyte-derived DCs from healthy individuals were generated without (control) or with IFN-λ (combination of IL-29 and IL-28A, called IFN-λ-DC) for 7 days, after which they were included in the mixed lymphocyte reaction (MLR) with allogeneic CD4⁺ T cells for 4 days. Co-culture supernatants from indicated groups were analyzed for IL-2 (A), IL-12 (B) by ELISA; mean±SD pg/ml shown. Neutralizing α-IL-10 or α-PD-1 antibodies, or recombinant (r) IL-2 or IL-12 were added during MLR and T cell proliferation was analyzed as described in Methods; shown mean±SD cpm from n = 5 sets of experiments (C). Entire MLR co-culture was analyzed for PD-L1(D) or PD-1 (E) mRNA. DCs were analyzed for PD-L1 expression by flow cytometry using isotype or specific fluorescent antibodies (MFI control 414±126 vs 622±98 IFN- λ DCs) (F, left). T cells from MLR co-culture cells were analyzed for PD-1 expression by flow cytometry using isotype or specific fluorescent antibodies (MFI control 125±27 vs 187±39 IFN- λ DCs) (F, right). * indicates p<0.05 compared to controls.

Figure 4. IFN-λ-DCs promote proliferation of existing but not de novo generation of regulatory T cells.

(A,B) CD4⁺ T cells were labeled with CFSE and cultured alone (red) or co-cultured with control (blue) or IFN-λ-DCs (green); at the end of culture CD4⁺CD25⁺ cells (A) or CD4⁺CD25⁻ cells (B) were analyzed for CFSE content by flow cytometry; representative histogram of n = 8 is shown. (C,D) The DC/T cells co-culture was analyzed for FoxP3 mRNA (C) using specific primers in quantitative PCR or for the frequency of FoxP3⁺ cells by flow cytometry (D); data are shown as fold change compared to control DC/T cells co-culture (C) or as representative flow cytometry dot blots (D). (E) T cells were included into 2 rounds of MLR. The first MLR was set up as described in Fig. 3 with CD4⁺ T cells as responders and control or IFN- λ-exposed DCs as stimulators. At the end of this first MLR, the T cells were purified into CD4⁺CD25⁻ and CD4⁺CD25⁺ populations using magnetic beads and equal numbers of cell types were included into a second round MLR with allogeneic normal DCs at indicated DC/T cell ratio. Proliferation during the second MLR was assessed by ³H-Td incorporation during the final 16 h of the 5-day co-culture. Mean±SD cpm values from n = 8 are shown. (F) IFN- λ-exposed DCs were co-cultured with CD4⁺ (including CD25⁺ and CD25^- populations) or CD4⁺CD25⁻ or CD4⁺CD25⁺ T cells; T cell cultures without DCs were used as controls. The co-cultures were analyzed by flow cytometry for the frequency of FoxP3⁺ cells (F). Representative set of flow cytometry dotblots of n = 4 with similar results is shown. * indicates p<0.05 compared to controls.